Get feedback from user and save image

README 3.md

@@ -1,12 +0,0 @@
----
-title: Ofwat Defects Classification Demo
-emoji: 😻
-colorFrom: gray
-colorTo: red
-sdk: gradio
-sdk_version: 4.7.1
-app_file: app.py
-pinned: false
----
-
-Check out the configuration reference at https://huggingface.co/docs/hub/spaces-config-reference

app.py

@@ -2,12 +2,45 @@ import gradio as gr
 from transformers import pipeline
 from PIL import Image
 import os
+from huggingface_hub import HfApi, upload_file
+import io
+import numpy as np
+import uuid
 
 # Initialize the pipeline with your model
 pipe = pipeline("image-classification", model="SubterraAI/ofwat_cleaner_classification")
-HF_TOKEN = os.environ.get('HF_TOKEN')
-print(HF_TOKEN)
-hf_writer = gr.HuggingFaceDatasetSaver(HF_TOKEN, "SubterraAI/ofwat_cleaner_loop", separate_dirs=True)
+HF_TOKEN = os.getenv('HF_TOKEN')
+DATASET_NAME = "SubterraAI/ofwat_cleaner_loop"
+hf_api = HfApi()
+
+# Directory where the flagged images will be saved
+flagged_data_dir = "./flagged_data"
+
+def simple_flag(image, label):
+    # Convert the input image to PIL format and save to a BytesIO object
+    pil_image = Image.fromarray(image.astype(np.uint8))
+    img_byte_arr = io.BytesIO()
+    pil_image.save(img_byte_arr, format='PNG')
+    
+    # Generate a unique ID for the image
+    unique_id = str(uuid.uuid4())
+    img_filename = f"{unique_id}.png"
+
+    # Save the image to a BytesIO object
+    image_bytes = img_byte_arr.getvalue()
+
+    # Upload the image to the correct label directory in the Hugging Face dataset
+    label_dir = f"{label}/{img_filename}"
+    upload_file(
+        path_or_fileobj=io.BytesIO(image_bytes),
+        path_in_repo=label_dir,
+        repo_id=DATASET_NAME,
+        repo_type="dataset",
+        token=HF_TOKEN,
+        commit_message=f"Add image with label {label}"
+    )
+
+    return "Image saved successfully to Hugging Face dataset."
 
 def classify_image(image):
     # Convert the input image to PIL format
@@ -19,23 +52,41 @@ def classify_image(image):
     # Extract labels and scores
     return {dic["label"]: dic["score"] for dic in res}
 
-# Create the Gradio interface
-iface = gr.Interface(
-    classify_image, 
-    "image", 
-    "label", 
-    examples=[
-        ["examples/CS.jpg"], 
-        ["examples/GI.jpg"],
-        ["examples/PP.jpg"]
-    ],
-    description="Upload an image to view a classification demonstration leveraging the dataset/library of images collected by WRc & Unitied Utitlies during The Water Services Regulation Authority (OFWAT) Innovation Challenge – Artificial Intelligence and Sewers. Not only can you see the initial classification, but you as the user can also inform us if the classification is correct. Your response will be used to retrain this model. The team at Subterra would like to thank all of those involved in collecting this dataset as we hope that other groups will use it to further advance technology solutions for the water industry.",
-    title="Sewer Obstruction Classification with AI by Subterra",
-    allow_flagging="manual",
-    flagging_options=["obstruction", "no_obstruction"],
-    flagging_callback=hf_writer
-
-)
-
-# Launch the interface
-iface.launch()
+def save_flagged_image(image, label):
+    try:
+        # Convert the input image to PIL format
+        PIL_image = Image.fromarray(image).convert('RGB')
+        
+        # Create the directory for the label if it doesn't exist
+        label_dir = os.path.join(flagged_data_dir, label)
+        os.makedirs(label_dir, exist_ok=True)
+        
+        # Save the image with a unique filename
+        img_filename = f"{label}_{hash(image.tobytes())}.png"
+        img_path = os.path.join(label_dir, img_filename)
+        PIL_image.save(img_path)
+        
+        return "Image saved successfully."
+    except Exception as e:
+        print(f"Error during flagging: {e}")
+        return f"An error occurred: {e}"
+
+with gr.Blocks() as demo:
+    gr.Markdown("# Sewer Obstruction Classification with AI by Subterra")
+    gr.Markdown("Upload an image to view a classification demonstration leveraging the dataset/library of images collected by WRc & United Utilities during The Water Services Regulation Authority (OFWAT) Innovation Challenge – Artificial Intelligence and Sewers. Not only can you see the initial classification, but you as the user can also inform us if the classification is correct. Your response will be used to retrain this model. The team at Subterra would like to thank all of those involved in collecting this dataset as we hope that other groups will use it to further advance technology solutions for the water industry.")
+
+    with gr.Row():
+        with gr.Column():
+            img_input = gr.Image()
+            submit_button = gr.Button("Classify")
+            examples = gr.Examples(["examples/CS.jpg", "examples/GI.jpg", "examples/PP.jpg"], inputs=img_input)
+        with gr.Column():
+            output_label = gr.Label()
+            flagging_options = gr.Radio(["obstruction", "no_obstruction"])
+            flag_button = gr.Button("Flag")
+            flag_status = gr.Textbox(visible=True)
+
+    submit_button.click(classify_image, inputs=img_input, outputs=output_label)
+    flag_button.click(simple_flag, inputs=[img_input, flagging_options], outputs=flag_status)
+
+demo.launch()

app_save.py

@@ -1,50 +0,0 @@
-import gradio as gr
-from transformers import pipeline
-from PIL import Image
-import os
-
-# Initialize the pipeline with your model
-pipe = pipeline("image-classification", model="SubterraAI/ofwat_cleaner_classification")
-HF_TOKEN = os.environ.get('HF_TOKEN')
-hf_writer = gr.HuggingFaceDatasetSaver(HF_TOKEN, dataset_name="ofwat_cleaner_loop", private=True, separate_dirs=True)
-
-def classify_image(image):
-    # Convert the input image to PIL format
-    PIL_image = Image.fromarray(image).convert('RGB')
-    
-    # Classify the image using the pipeline
-    res = pipe(PIL_image)
-    
-    # Extract labels and scores
-    return {dic["label"]: dic["score"] for dic in res}
-
-def flag_feedback(image, option, flag_status):
-    # Perform flagging action here using hf_writer
-    hf_writer.flag((image, option))
-
-    # Update the flag status to indicate feedback has been submitted
-    flag_status.update("Feedback submitted. Thank you!")
-    return flag_status
-
-# Create a state variable for the flag status
-flag_status = gr.State("")
-
-# Create the Gradio interface
-iface = gr.Interface(
-    classify_image,
-    inputs=[gr.Image(), gr.Radio(["obstruction", "no_obstruction"])],
-    outputs=[gr.Label(), gr.Textbox(label="Flag Status", value=flag_status)],
-    examples=[
-        ["examples/CS.jpg"],
-        ["examples/GI.jpg"],
-        ["examples/PP.jpg"]
-    ],
-    description="Upload an image to view a classification demonstration...",
-    title="Sewer Obstruction Classification with AI by Subterra",
-    allow_flagging="manual",
-    flagging_options=["obstruction", "no_obstruction"],
-    flagging_callback=lambda image, option: flag_feedback(image, option, flag_status)
-)
-
-# Launch the interface
-iface.launch()

flagged/log.csv

@@ -1,8 +0,0 @@
-image,output 0,output 1,flag,username,timestamp
-"{""path"":""flagged/image/d97b9ae054eab1bd10dc/PP.jpg"",""url"":""http://127.0.0.1:7860/file=/private/var/folders/5q/yl8pmxm116g6r3k8fd9gk74m0000gn/T/gradio/12885a897d79141c54ccb542582449159242eaa0/PP.jpg"",""size"":null,""orig_name"":""PP.jpg"",""mime_type"":null}","{""label"":""VC"",""confidences"":[{""label"":""VC"",""confidence"":0.7706952691078186},{""label"":""PVC"",""confidence"":0.20648105442523956},{""label"":""PE"",""confidence"":0.005161861423403025},{""label"":""PF"",""confidence"":0.004196972120553255},{""label"":""CO"",""confidence"":0.0034184197429567575}]}","| Prefix | Full Name |
-| ------ | --------- |
-| CS | Carbon Steel |
-| GI | Galvanized Iron |
-| PP | Polypropylene |
-| RC | Reinforced Concrete |",,,2023-11-27 15:26:38.548865
-"{""path"":""flagged/image/f38c599fa2de8d03404c/RC.jpg"",""url"":""http://127.0.0.1:7860/file=/private/var/folders/5q/yl8pmxm116g6r3k8fd9gk74m0000gn/T/gradio/6ab5f5d1df06303f856175903bbfc4639d4780b3/RC.jpg"",""size"":null,""orig_name"":""RC.jpg"",""mime_type"":null}","{""label"":""obstruction"",""confidences"":[{""label"":""obstruction"",""confidence"":0.988347589969635},{""label"":""no_obstruction"",""confidence"":0.01165243424475193}]}",obstruction,,2024-01-09 12:23:43.084607

unit_test.py

@@ -0,0 +1,11 @@
+import gradio as gr
+import os
+HF_TOKEN = os.environ.get('HF_TOKEN')
+hf_writer = gr.HuggingFaceDatasetSaver(HF_TOKEN, "image-classification-mistakes")
+def image_classifier(inp):
+    return {'cat': 0.3, 'dog': 0.7}
+demo = gr.Interface(fn=image_classifier, inputs="image", outputs="label",
+                allow_flagging="manual", flagging_callback=hf_writer)
+
+# Launch the interface
+demo.launch()